CN109345381A - A kind of Risk Identification Method and system - Google Patents
A kind of Risk Identification Method and system Download PDFInfo
- Publication number
- CN109345381A CN109345381A CN201811558708.3A CN201811558708A CN109345381A CN 109345381 A CN109345381 A CN 109345381A CN 201811558708 A CN201811558708 A CN 201811558708A CN 109345381 A CN109345381 A CN 109345381A
- Authority
- CN
- China
- Prior art keywords
- promise
- enterprise
- risk
- breaking
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- Finance (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Accounting & Taxation (AREA)
- Educational Administration (AREA)
- Technology Law (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
This application provides a kind of Risk Identification Method, for being identified to enterprise's risk of breaking one's promise, which comprises the break one's promise test data of risk of the enterprise is input to decision-tree model, for the decision tree learning to generate risk identification model of breaking one's promise；The enterprise is obtained to break one's promise the data to be predicted of risk；The break one's promise data to be predicted of risk of the enterprise are input in the risk identification model of breaking one's promise, are predicted by the risk identification model of breaking one's promise based on the break one's promise data to be predicted of risk of the enterprise；The prediction result is shown in the display interface of the terminal, so that user can clearly know prediction result.Present invention also provides a kind of risk recognition systems.By the Risk Identification Method and system, solve it is existing can not comprehensive and accurate analysis order of magnitude big business data by artificial mode.
Description
Technical field
The application belongs to technical field of data processing, and in particular to a kind of Risk Identification Method and system.
Background technique
With internet and economic continuous development, the emergence of internet finance, the consumer finance also promotes China's credit
The fast development of industry.Enterprise needs to realize the preparation of fund by the financial instrument in financial market in production management process
Deng causing to bring more credit risks.Currently, the financial sub-control of the traditional forms of enterprises is submitted to audit by manually carrying out, from data, need
Plenty of time and manpower and material resources are wanted, but also is easy to appear careless mistake, is also come into being based on this credit scoring technology, by loan
The integration of Kuan Ren each side data information predicts its credit analysis value, to help credit approval person to make a policy.
Due to the problems such as business data dimension is high and business data amount is big, traditional artificial there is presently no preferable sides
Method, which is realized, carries out effective forecast analysis to business risk, for potential risk existing for enterprise itself, affiliated enterprise of enterprise without
Method gives warning in advance, and enterprise can not accomplish effective risk averse, this is likely to result in the bad development of enterprise.
Summary of the invention
In order to solve the above problems existing in the present technology, the application is designed to provide a kind of Risk Identification Method and is
System, it is intended to solve the problems, such as that existing artificial treatment can not carry out effectively analysis prediction to order of magnitude big business information.
In order to solve the above technical problems, being applied to terminal this application provides a kind of Risk Identification Method, for enterprise
Risk of breaking one's promise is identified, which comprises the break one's promise test data of risk of the enterprise is input to decision-tree model, is supplied
The decision tree learning is to generate risk identification model of breaking one's promise；The enterprise is obtained to break one's promise the data to be predicted of risk；It will be described
The break one's promise data to be predicted of risk of enterprise are input in the risk identification model of breaking one's promise, and pass through the risk identification model of breaking one's promise
It is predicted based on the break one's promise data to be predicted of risk of the enterprise；The prediction result is shown in display circle of the terminal
Face, so that user can clearly know prediction result.
Optionally, the test data includes break one's promise enterprise and number of the enterprise of not breaking one's promise, wherein in the test data
The quantity ratio of break one's promise enterprise and the number of the enterprise of not breaking one's promise is 1:2.
Optionally, the break one's promise test data of risk of the enterprise is input to decision-tree model, for the decision tree learning
With generate break one's promise risk identification model the step of, comprising: the test data is standardized；By to the number after standardization
According to carry out cleaning and selection obtain characteristic；The characteristic is input to decision-tree model, for the decision tree learning
To generate risk identification model of breaking one's promise.
Optionally, the characteristic includes but is not limited to industrial and commercial basic modification information, trade information, judgement document time
The enterprise that the feature for the enterprise that number, enterprise shareholder externally hold a post, business entity externally hold a post revokes, as judge's text of defendant
Book number.
Optionally, described that the characteristic is input to decision-tree model, it is broken one's promise for the decision tree learning with generating
The step of risk identification model, comprising: current optimal characteristics are determined according to preset feature selecting algorithm and to disruptive features number
According to；Decision tree present node is established according to the current optimal characteristics, is worked as according to described established to disruptive features data with described
Front nodal point is the branch of father node, is directly only satisfied with preset condition stopping and continues to construct decision tree to generate risk identification mould of breaking one's promise
Type.
Optionally, the decision-tree model is CART decision Tree algorithms, described to be determined according to preset feature selecting algorithm
It is current optimal characteristics that current optimal characteristics, which are the maximum feature of determining information gain,.
Optionally, described straight to be only satisfied with preset condition stopping and continue to construct decision tree to generate risk identification model of breaking one's promise
The step of, comprising: construct complete initial decision tree；Rear pruning algorithms are used to handle to obtain the initial decision tree
Risk identification of breaking one's promise model.
Optionally, described that rear pruning algorithms is used to be handled the initial decision tree with the risk identification mould that obtains breaking one's promise
The step of type, comprising: the verify data for breaking one's promise risk identification model is input to initial decision tree；According to lower error rate
Pruning algorithms handle the initial decision tree, to obtain final risk identification model of breaking one's promise.
Optionally, the verify data includes break one's promise enterprise and number of the enterprise of not breaking one's promise, wherein in the verify data
The quantity ratio of break one's promise enterprise and the number of the enterprise of not breaking one's promise is 7:3.
Present invention also provides a kind of risk recognition system, the risk recognition system is used for risk progress of breaking one's promise enterprise
Identification, the system comprises achievement module, for the break one's promise test data of risk of the enterprise to be input to decision-tree model,
For the decision tree learning to generate risk identification model of breaking one's promise；Obtain module, for obtain the enterprise break one's promise risk to
Prediction data；Prediction module, for the break one's promise data to be predicted of risk of the enterprise to be input to the risk identification mould of breaking one's promise
In type, predicted by the risk identification model of breaking one's promise based on the break one's promise data to be predicted of risk of the enterprise；Show mould
Block, for the prediction result to be shown in the display interface of the terminal, so that user can clearly know prediction result.
The application is by from existing company information data, by CART decision Tree algorithms in company information data
The data characteristics of various aspects is learnt, and is up to standard with information gain to select optimal characteristics to carry out division achievement, is led to
Exist after first establishing complete decision tree, then the initial decision tree is handled using lower error rate pruning algorithms, with
To final risk identification model of breaking one's promise.In this manner, the characteristic for making the branch node of tree be included as far as possible to the greatest extent may be used
Same category can be belonged to, i.e. " purity " of node is higher and higher, becomes unordered characteristic orderly, and then the risk is known
Other model is applied in data to be predicted with the risk of breaking one's promise of enterprise corresponding to determination data to be predicted, is solved existing logical
Crossing artificial mode can not comprehensive and accurate analysis order of magnitude big business data.
Detailed description of the invention
Fig. 1 is the application flow chart.
Fig. 2 is the application decision tree flow chart.
Specific embodiment
In order to make the above objects, features, and advantages of the present application more apparent, with reference to the accompanying drawing and it is specific real
Applying mode, the present application will be further described in detail.
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
In subsequent description, it is only using the suffix for indicating such as " module ", " component " or " unit " of element
Be conducive to explanation of the invention, itself there is no a specific meaning.Therefore, " module ", " component " or " unit " can mix
Ground uses.
Fig. 1 is the flow chart of a Risk Identification Method provided by the present application.The method of the embodiment is once triggered by user,
Then the process in the embodiment passes through terminal automatic running, wherein each step can be when operation according to such as process
Sequence in figure successively carries out, and is also possible to multiple steps according to the actual situation while carrying out, herein and without limitation.The application
The Risk Identification Method of offer is used to identify enterprise's risk of breaking one's promise.Information cuing method provided by the present application includes as follows
Step:
The break one's promise test data of risk of the enterprise is input to decision-tree model, learned for the decision tree by step S110
It practises to generate risk identification model of breaking one's promise；
Step S120 obtains the enterprise and breaks one's promise the data to be predicted of risk；
The break one's promise data to be predicted of risk of the enterprise are input in the risk identification model of breaking one's promise by step S130,
It is predicted by the risk identification model of breaking one's promise based on the break one's promise data to be predicted of risk of the enterprise；
The prediction result is shown in the display interface of the terminal by step S140, so that user can clearly obtain
Know prediction result.
By Risk Identification Method provided by the present application, become unordered characteristic orderly, and then the risk is known
Other model is applied in data to be predicted with the risk of breaking one's promise of enterprise corresponding to determination data to be predicted, is solved existing logical
Crossing artificial mode can not comprehensive and accurate analysis order of magnitude big business data.
Detailed narration is carried out to above steps below in conjunction with specific embodiment.
In step s 110, the break one's promise test data of risk of the enterprise is input to decision-tree model, for the decision
Tree study is to generate risk identification model of breaking one's promise.
Specifically, sample data is obtained in advance, and sample data is carried out to be classified as test data and verify data.Wherein,
Test data for machine learning model for being learnt to establish complete decision tree.Verify data is logical for being applied to foundation
The decision tree for crossing test data foundation carries out beta pruning, is examined with providing protection to the false feature in overfitting test data.
In the present embodiment, the test data includes break one's promise enterprise and number of the enterprise of not breaking one's promise, wherein in the test data
The quantity ratio of break one's promise enterprise and the number of the enterprise of not breaking one's promise is 1:2, and for example, number of the enterprise of breaking one's promise is 504397.
In the present embodiment, step S110 can be achieved by the steps of:
Step S1101 is standardized the test data；
Step S1102, by the data after standardization carry out cleaning and selection obtain characteristic；
The characteristic is input to decision-tree model, for the decision tree learning to generate wind of breaking one's promise by step S1103
Dangerous identification model.
Specifically, due in test data include lteral data and numeric data, in step S1101, need
The numeric data that algorithm can identify is converted by categorical datas such as test data lteral datas.Original training data is carried out
Standardization, for example, carries out numerical value conversion for text variable by way of dictionary.
In step S1102, by being cleaned initial data to obtain characteristic and flag data, then pass through
Characteristic and labeled data are handled, such as specimen sample, sample tune power, abnormal point removal, feature normalization processing
Deng.In the present embodiment, risk identification model is trained using following characteristic, wherein characteristic includes but not
It is limited to: the feature for the enterprise that industrial and commercial basic modification information, trade information, judgement document's number, enterprise shareholder externally hold a post, enterprise
The enterprise that legal person externally holds a post revokes, as judgement document's number of defendant.Wherein, industrial and commercial basic modification information can wrap
It includes: setting up the time limit, registered capital change number, address change number, shareholder's change number, alteration of judicial person number etc..Industry letter
Breath may include industry and enterprise quantity, industry and enterprise revokes quantity, industry and enterprise revokes rate etc..The enterprise that enterprise shareholder externally holds a post
The feature of industry, which may include, revokes, executes, breaking one's promise, as judgement document's number of defendant etc..
In step S1103, decision-tree model refers to the basic algorithm for establishing tree, by the way that test data to be applied to
The basic algorithm is to obtain final decision tree.In the present embodiment, the decision-tree model is CART decision Tree algorithms.It needs
It is noted that can also be realized using other achievement algorithms, specifically without limitation.Specifically, in the present embodiment,
Step S1103 includes the following steps:
Step S11031 determines current optimal characteristics according to preset feature selecting algorithm and to disruptive features data；
Step S11032 establishes decision tree present node according to the current optimal characteristics, according to described to disruptive features
Data are established using the present node as the branch of father node, are directly only satisfied with preset condition stopping and are continued to construct decision tree with life
At risk identification model of breaking one's promise.
Specifically, in the present embodiment, described to determine that current optimal characteristics are true according to preset feature selecting algorithm
Determining the maximum feature of information gain is current optimal characteristics.In step S11031, by comentropy and information gain ratio come into
The feature selecting of row risk identification model, according to determining optimal characteristics and to disruptive features data.Wherein, optimal characteristics are used for
Current spliting node is established, to disruptive features data for the branch as preceding spliting node, and continues through comentropy and letter
The ratio of gains is ceased to determine the optimal characteristics divided next time.
Firstly, determining whether that risky is label, the information of label is calculated according to the sample data in characteristic
Entropy.Wherein, source signal has n kind value: U1,U2,...,Un, corresponding probability is P1,P2,...,Pn, and various symbols occur
It is independent, then the uncertainty-logP of the average uncertain single symbol of the information sourceiAssembly average E, as information
Entropy indicates are as follows:
Secondly, with currently for the feature X of root node, entropy, that is, H (D | X) of remaining characteristic is calculated, is subtracted according to original entropy
Remove remaining entropy after disruptive features and obtains information gain g (D, X)=H (D)-H (D | X)；Again using other characteristic Ys as root section
Information gain g (D, Y)=H (D)-H (D | Y) is calculated in point；Similarly corresponding information is obtained again with other feature calculations to increase
Benefit.
Finally, determining that maximum information yield value institute is right by comparing the numerical values recited of g (D, X), g (D, Y) ... g (D, N)
The feature answered is optimal characteristics, and then other features are to disruptive features data.
In step S11032, after determining present node, calculate separately when present node takes "No", corresponding label
Information gain A；With when present node takes "Yes", to the maximum information gain B in disruptive features data.Then comparison information
Gain A and information gain B determine next segmentation direction according to size.Each node is counted respectively according to above-mentioned steps
Point counting is cut, and the sample for making the branch node of tree be included belongs to same category as far as possible, i.e. " purity " (purity) of node is more
Come higher, unordered data is made to become orderly.
For example, by taking characteristic includes following content as an example: judgement document's number, alteration of judicial person number, network
Execution number, the industry and enterprise of shareholder or investments abroad enterprise revoke quantity.Risk is established according to method as described above to know
Other model is as follows:
In above-mentioned { } braces, it is a series of outputs that the decision tree based on generation obtains, includes decision_
Path, that is, decision path also includes the Covering samples number sample_number of current leaf node, and current decision path point
Class accounting that maximum a kind of label label, the false segmentation rate missclassfication_rate calculated based on this；It would generally
The decision path that those Covering samples numbers are more and false segmentation rate is smaller is selected to be predicted as final decision rule.See figure
2。
Wherein, samples indicates that the sample number under the conditions of present node, error_rate indicate that mistake divides rate inside box,
Wherein mistake point rate is to be based on: the affiliated label under the conditions of present node is classified as the most label of Covering samples amount, error rate
It is exactly that those are based on those of labeling mistake sample, its calculation formula is: 1-max p (i) | and i in [1:n] }, herein
On the basis of, if there is multiple attributes, can also continue to separate attribute, the growth set in the manner described above, until growing up to one
Complete tree.
In the present embodiment, step S11032 further include: construct complete initial decision tree；Using rear pruning algorithms pair
The initial decision tree is handled with the risk identification model that obtains breaking one's promise.
In the present embodiment, it uses rear pruning algorithms to handle the initial decision tree and is known with the risk that obtains breaking one's promise
The step of other model, which can be accomplished in that, is input to initial determine for the verify data for breaking one's promise risk identification model
Plan tree；The initial decision tree is handled according to lower error rate pruning algorithms, to obtain final risk identification of breaking one's promise
Model.In the present embodiment, the verify data includes break one's promise enterprise and number of the enterprise of not breaking one's promise, wherein the verifying number
The quantity ratio of enterprise and number of the enterprise of not breaking one's promise of breaking one's promise described in is 7:3.In other embodiments, rear pruning algorithms are also
It can be pessimistic wrong pruning algorithms, cost complexity pruning algorithms, based on wrong pruning algorithms etc., specifically without limitation.
By above embodiment, risk identification model can establish, can solve by the risk identification model existing
Some can not comprehensive and accurate analysis order of magnitude big business data by artificial mode.
In the step s 120, the enterprise is obtained to break one's promise the data to be predicted of risk.
In the present embodiment, enterprise to be predicted refers to the enterprise of its business risk to be assessed.Data to be predicted include but
It is not limited to feature, the enterprise of the enterprise that industrial and commercial modification information, trade information, judgement document's number, enterprise shareholder substantially externally hold a post
The enterprise that industry legal person externally holds a post revokes, as judgement document's number of defendant.
Specifically, in the present embodiment, data to be predicted can be preset, then by way of web crawlers
Data relevant to the data to be predicted are obtained from network automatically by web crawlers.In other embodiments, it is also possible to
Relative data are actively supplied to the user using Risk Identification Method provided by the present application by enterprise to be predicted, are used
The information data that person is provided using enterprise to be predicted is completed risk identification by following step and is predicted.
In step s 130, the break one's promise data to be predicted of risk of the enterprise are input to the risk identification model of breaking one's promise
In, it is predicted by the risk identification model of breaking one's promise based on the break one's promise data to be predicted of risk of the enterprise.
In step S140, the prediction result is shown in the display interface of the terminal, so that user can understand
Know prediction result in ground.
Pass through above embodiment, it is possible to reduce data normalization process, lift scheme establishes speed, and is being built
During tree, more errors can be reduced, obtain better precision.Meanwhile it being used during risk identification model learning
A variety of different type company information datas, so that the model established is more perfect, to ensure the accuracy of risk profile.
The application also provides a kind of risk recognition system, and the risk recognition system risk that is used to break one's promise to enterprise is known
Not, the system comprises:
Achievement module, for the break one's promise test data of risk of the enterprise to be input to decision-tree model, for the decision
Tree study is to generate risk identification model of breaking one's promise；
Module is obtained, is broken one's promise the data to be predicted of risk for obtaining the enterprise；
Prediction module, for the break one's promise data to be predicted of risk of the enterprise to be input to the risk identification model of breaking one's promise
In, it is predicted by the risk identification model of breaking one's promise based on the break one's promise data to be predicted of risk of the enterprise；
Display module, for the prediction result to be shown in the display interface of the terminal, so that user can understand
Know prediction result in ground.
Optionally, the achievement model is also used to be standardized the test data；By to the number after standardization
According to carry out cleaning and selection obtain characteristic；The characteristic is input to decision-tree model, for the decision tree learning
To generate risk identification model of breaking one's promise.
Optionally, the achievement model, be also used to be determined according to preset feature selecting algorithm current optimal characteristics and to
Disruptive features data；Decision tree present node is established according to the current optimal characteristics, is built according to described to disruptive features data
It is vertical using the present node as the branch of father node, it is straight be only satisfied with preset condition stopping and continue to construct decision tree broken one's promise with generating
Risk identification model.
Optionally, the achievement model is also used to construct complete initial decision tree；Using rear pruning algorithms to described first
Beginning decision tree is handled with the risk identification model that obtains breaking one's promise.
Optionally, the achievement model is also used to for the verify data for breaking one's promise risk identification model being input to initially
Decision tree；The initial decision tree is handled according to lower error rate pruning algorithms, is known with the risk of breaking one's promise for obtaining final
Other model.
It should be noted that the content in systems approach embodiment equally can be using in method implementation above-mentioned
Content, therefore, this will not be repeated here.
The application is not limited to above-mentioned optional embodiment, anyone can show that other are various under the enlightenment of the application
The product of form, however, make any variation in its shape or structure, it is all to fall into the claim of this application confining spectrum
Technical solution, all fall within the protection scope of the application.
Claims (10)
1. a kind of Risk Identification Method, which is characterized in that it is applied to terminal, it is described for being identified to enterprise's risk of breaking one's promise
Method includes:
The break one's promise test data of risk of the enterprise is input to decision-tree model, for the decision tree learning to generate wind of breaking one's promise
Dangerous identification model；
The enterprise is obtained to break one's promise the data to be predicted of risk；
The break one's promise data to be predicted of risk of the enterprise are input in the risk identification model of breaking one's promise, the wind of breaking one's promise is passed through
Dangerous identification model is predicted based on the break one's promise data to be predicted of risk of the enterprise；
The prediction result is shown in the display interface of the terminal, so that user can clearly know prediction result.
2. Risk Identification Method as described in claim 1, which is characterized in that the test data includes breaking one's promise enterprise and not lose
Believe number of the enterprise, wherein the break one's promise quantity ratio of enterprise and number of the enterprise of not breaking one's promise of described in the test data is 1:2.
3. Risk Identification Method as described in claim 1, which is characterized in that by the enterprise break one's promise risk test data it is defeated
Enter to decision-tree model, broken one's promise for the decision tree learning with generating risk identification model the step of, comprising:
The test data is standardized；
By to the data after standardization carry out cleaning and selection obtain characteristic；
The characteristic is input to decision-tree model, for the decision tree learning to generate risk identification model of breaking one's promise.
4. Risk Identification Method as claimed in claim 3, which is characterized in that the characteristic includes but is not limited to industrial and commercial base
The feature of the enterprise that this modification information, trade information, judgement document's number, enterprise shareholder externally hold a post, business entity are to local official
The enterprise of duty revokes, as judgement document's number of defendant.
5. Risk Identification Method as described in claim 1, which is characterized in that described that the characteristic is input to decision tree
Model, broken one's promise for the decision tree learning with generating risk identification model the step of, comprising:
Current optimal characteristics are determined according to preset feature selecting algorithm and to disruptive features data；
Decision tree present node is established according to the current optimal characteristics, is worked as according to described established to disruptive features data with described
Front nodal point is the branch of father node, is directly only satisfied with preset condition stopping and continues to construct decision tree to generate risk identification mould of breaking one's promise
Type.
6. Risk Identification Method as claimed in claim 5, which is characterized in that the decision-tree model is the calculation of CART decision tree
Method, it is described according to preset feature selecting algorithm determine current optimal characteristics be the maximum feature of determining information gain be it is current most
Excellent feature.
7. Risk Identification Method as claimed in claim 5, feature is being, it is described it is straight be only satisfied with preset condition stop after
Continuous building decision tree broken one's promise with generating risk identification model the step of, comprising:
Construct complete initial decision tree；
Rear pruning algorithms are used to be handled the initial decision tree with the risk identification model that obtains breaking one's promise.
8. Risk Identification Method as described in claim 1, which is characterized in that described initially to be determined using rear pruning algorithms to described
Plan tree handled with obtain breaking one's promise risk identification model the step of, comprising:
The verify data for breaking one's promise risk identification model is input to initial decision tree；
The initial decision tree is handled according to lower error rate pruning algorithms, to obtain final risk identification mould of breaking one's promise
Type.
9. Risk Identification Method as claimed in claim 8, which is characterized in that the verify data includes breaking one's promise enterprise and not lose
Believe number of the enterprise, wherein the break one's promise quantity ratio of enterprise and number of the enterprise of not breaking one's promise of described in the verify data is 7:3.
10. a kind of risk recognition system, which is characterized in that the risk recognition system is used to know enterprise's risk of breaking one's promise
Not, the system comprises:
Achievement module is learned for the break one's promise test data of risk of the enterprise to be input to decision-tree model for the decision tree
It practises to generate risk identification model of breaking one's promise；
Module is obtained, is broken one's promise the data to be predicted of risk for obtaining the enterprise；
Prediction module, for the break one's promise data to be predicted of risk of the enterprise to be input in the risk identification model of breaking one's promise,
It is predicted by the risk identification model of breaking one's promise based on the break one's promise data to be predicted of risk of the enterprise；
Display module, for the prediction result to be shown in the display interface of the terminal, so that user can clearly obtain
Know prediction result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811558708.3A CN109345381A (en) | 2018-12-19 | 2018-12-19 | A kind of Risk Identification Method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811558708.3A CN109345381A (en) | 2018-12-19 | 2018-12-19 | A kind of Risk Identification Method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109345381A true CN109345381A (en) | 2019-02-15 |
Family
ID=65304569
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811558708.3A Pending CN109345381A (en) | 2018-12-19 | 2018-12-19 | A kind of Risk Identification Method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109345381A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110033159A (en) * | 2019-02-27 | 2019-07-19 | 阿里巴巴集团控股有限公司 | Risk Identification Method, device and equipment |
| CN110163481A (en) * | 2019-04-19 | 2019-08-23 | 深圳壹账通智能科技有限公司 | Electronic device, user's air control auditing system test method and storage medium |
| CN112785427A (en) * | 2021-03-15 | 2021-05-11 | 国网青海省电力公司西宁供电公司 | Enterprise credit analysis system based on electric power data |
| CN113313417A (en) * | 2021-06-23 | 2021-08-27 | 北京鼎泰智源科技有限公司 | Complaint risk signal grading method and device based on decision tree model |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050273430A1 (en) * | 2004-06-02 | 2005-12-08 | Pliha Robert K | Systems and methods for scoring bank customers direct deposit account transaction activity to match financial behavior to specific acqusition, performance and risk events defined by the bank using a decision tree and stochastic process |
| CN101996381A (en) * | 2009-08-14 | 2011-03-30 | 中国工商银行股份有限公司 | Method and system for calculating retail asset risk |
| CN105590261A (en) * | 2014-12-31 | 2016-05-18 | 中国银联股份有限公司 | Merchant risk estimation method and system |
| CN107451710A (en) * | 2017-04-27 | 2017-12-08 | 北京鼎泰智源科技有限公司 | A kind of Information Risk grade five-category method and system |
| CN107920055A (en) * | 2017-09-27 | 2018-04-17 | 中国银联股份有限公司 | A kind of IP risk evaluating methods and IP Risk Evaluating Systems |
-
2018
- 2018-12-19 CN CN201811558708.3A patent/CN109345381A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050273430A1 (en) * | 2004-06-02 | 2005-12-08 | Pliha Robert K | Systems and methods for scoring bank customers direct deposit account transaction activity to match financial behavior to specific acqusition, performance and risk events defined by the bank using a decision tree and stochastic process |
| CN101996381A (en) * | 2009-08-14 | 2011-03-30 | 中国工商银行股份有限公司 | Method and system for calculating retail asset risk |
| CN105590261A (en) * | 2014-12-31 | 2016-05-18 | 中国银联股份有限公司 | Merchant risk estimation method and system |
| CN107451710A (en) * | 2017-04-27 | 2017-12-08 | 北京鼎泰智源科技有限公司 | A kind of Information Risk grade five-category method and system |
| CN107920055A (en) * | 2017-09-27 | 2018-04-17 | 中国银联股份有限公司 | A kind of IP risk evaluating methods and IP Risk Evaluating Systems |
Non-Patent Citations (1)
| Title |
|---|
| 柴明锐: "《数据挖掘技术及在石油地质中的应用》", 30 September 2017 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110033159A (en) * | 2019-02-27 | 2019-07-19 | 阿里巴巴集团控股有限公司 | Risk Identification Method, device and equipment |
| CN110163481A (en) * | 2019-04-19 | 2019-08-23 | 深圳壹账通智能科技有限公司 | Electronic device, user's air control auditing system test method and storage medium |
| CN112785427A (en) * | 2021-03-15 | 2021-05-11 | 国网青海省电力公司西宁供电公司 | Enterprise credit analysis system based on electric power data |
| CN112785427B (en) * | 2021-03-15 | 2024-04-26 | 国网青海省电力公司西宁供电公司 | Enterprise credit analysis system based on power data |
| CN113313417A (en) * | 2021-06-23 | 2021-08-27 | 北京鼎泰智源科技有限公司 | Complaint risk signal grading method and device based on decision tree model |
| CN113313417B (en) * | 2021-06-23 | 2024-01-26 | 北京鼎泰智源科技有限公司 | Method and device for classifying complaint risk signals based on decision tree model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109345381A (en) | A kind of Risk Identification Method and system | |
| CN107993139A (en) | A kind of anti-fake system of consumer finance based on dynamic regulation database and method | |
| CN110415111A (en) | Merge the method for logistic regression credit examination & approval with expert features based on user data | |
| CN109657977A (en) | A kind of Risk Identification Method and system | |
| CN108845988B (en) | Entity identification method, device, equipment and computer readable storage medium | |
| CN107704512A (en) | Financial product based on social data recommends method, electronic installation and medium | |
| CN109859052A (en) | A kind of intelligent recommendation method, apparatus, storage medium and the server of investment tactics | |
| CN106294762A (en) | A kind of entity recognition method based on study | |
| Gu et al. | Application of fuzzy decision tree algorithm based on mobile computing in sports fitness member management | |
| CN108241872A (en) | The adaptive Prediction of Stock Index method of Hidden Markov Model based on the multiple features factor | |
| CN113590698A (en) | Artificial intelligence technology-based data asset classification modeling and hierarchical protection method | |
| CN111738856A (en) | Stock public opinion investment decision analysis method and device | |
| CN110956541A (en) | Stock tendency classification prediction method based on intelligent fusion calculation | |
| CN110704803A (en) | Target object evaluation value calculation method and device, storage medium and electronic device | |
| CN113283673A (en) | Model performance attenuation evaluation method, model training method and device | |
| CN110807174B (en) | Effluent analysis and abnormity identification method for sewage plant group based on statistical distribution | |
| CN106156107A (en) | A kind of discovery method of hot news | |
| CN108711074B (en) | Service classification method, device, server and readable storage medium | |
| CN115936389A (en) | Big data technology-based method for matching evaluation experts with evaluation materials | |
| CN114529255A (en) | Loan automatic approval method and system based on wind control scoring card | |
| CN108363738A (en) | A kind of recommendation method of industrial equipment data analysis algorithm | |
| CN107967321A (en) | A kind of crop breeding evaluation method based on hierarchical support vector machines | |
| CN114880635A (en) | User security level identification method, system, electronic device and medium of model integrated with lifting tree construction | |
| Zhou et al. | Bank Customer Classification Algorithm Based on Improved Decision Tree | |
| CN116541527B (en) | Document classification method based on model integration and data expansion |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190215 |
|
| WD01 | Invention patent application deemed withdrawn after publication |